The problem of accurately and early predicting Parkinson's Disease (PD) remains a challenge, as traditional clinical assessments may not be sensitive enough to detect subtle markers of PD at an early stage. Existing prediction methods lack the precision, scalability, and interpretability needed for widespread clinical implementation. Therefore, there is a need to develop robust and interpretable machine learning (ML) models that can leverage diverse datasets to accurately predict PD, enabling early intervention strategies and improving patient outcomes.
The objective is to comprehensively review and analyze the current state-of-the-art in the development of machine learning (ML) models for prediction of Parkinson's Disease (PD) and provide an enhanced model for prediction of PD. This includes exploring various ML techniques, data types, and challenges associated with PD prediction.
The project aims to provide insights into the strengths, limitations, and future directions of ML-based PD prediction, and highlight opportunities for improving accuracy, interpretability, and clinical implementation of ML models for PD prediction.
kaggle link : https://www.kaggle.com/datasets/vikasukani/parkinsons-disease-data-set
Each row in the table represents one of the 195 voice recordings, and each column represents a specific voice measure (the "name" column). The "status" column, which is set to 0 for healthy and 1 for PD, is used to distinguish between healthy individuals and those with PD in the data.
1.Download the git repo
2.Change the relative path of the parkinsons.csv file in the .ipynb file according to your computer directory.
3.Run the file in your ide.It will take about 5-30 mins depending on your computer performance.